Radio frequency (RF) transmitters are found in many one-way and two-way communication devices, such as portable communication devices, (cellular telephones), personal digital assistants (PDAs) and other communication devices. An RF transmitter must transmit using whatever communication methodology is dictated by the particular communication system within which it is operating. For example, communication methodologies typically include amplitude modulation, frequency modulation, phase modulation, or a combination of these. In a typical global system for mobile communications (GSM) mobile communication system using narrowband TDMA technology, a Gaussian minimum shift keying (GMSK) modulation scheme supplies a clean phase-modulated (PM) transmit signal to a non-linear power amplifier directly from an oscillator.
In such an arrangement, a non-linear power amplifier, which is highly efficient, can be used, thus allowing efficient transmission of the phase-modulated signal and minimizing power consumption. Because the modulated signal is supplied directly from an oscillator, the need for filtering, either before or after the power amplifier, is minimized. Other transmission standards, such as that employed in IS-136, however, use a modulation scheme in which the transmitted signal is both phase modulated (PM) and amplitude modulated (AM). Standards such as these increase the data rate without increasing the bandwidth of the transmitted signal. Unfortunately, existing GSM transmitter hardware is not easily adapted to transmit a signal that includes both a PM component and an AM component. One reason for this difficulty is that in order to transmit a signal containing a PM component and an AM component, a highly linear power amplifier is required. Unfortunately, highly linear power amplifiers are very inefficient, thus consuming significantly more power than a non-linear power amplifier and drastically reducing the life of the battery or other power source.
This condition is further complicated because transmitters typically employed in GSM communication systems transmit in bursts and must be able to control the ramp-up of the transmit power as well as have a high degree of control over the output power level over a wide power range. In GSM this power control is typically performed using a closed feedback loop in which a portion of the signal output from the power amplifier is compared with a reference signal and the resulting error signal is fed back to the control port of the power amplifier.
The EDGE communication system attempts to increase the data transmission capability of a GSM communication system by including an amplitude modulation (AM) component in the transmit signal. However, when attempting to add an AM component to the GSM type modulation system, the power control loop will attenuate the amplitude variations present in the signal in an attempt to maintain a constant output power. In such an arrangement, the power control loop tends to cancel the AM portion of the signal.
Further, in those transmission standards in which both a PM signal and an AM signal are sent to a power amplifier, unless the power amplifier is very linear, it may distort the combined transmission signal by causing undesirable AM to PM conversion. This conversion is detrimental to the transmit signal and can require the use of a costly and inefficient linear power amplifier.
In the past, the transceiver components for such a communication system were typically implemented using multiple devices, also referred to as “chips.” However, industry pressures to reduce cost, implementation complexity and power consumption and to extend battery life are driving the industry to attempt single chip architectures. Unfortunately, a single chip implementation for a GSM/EDGE transceiver presents many design challenges, especially in a system in which a closed power control loop is used to control output power of the transmitter. For example, when a closed loop power control system is implemented on the same chip as the transceiver components, the radio frequency (RF) on-chip isolation requirement between the components becomes a major factor affecting transceiver performance.
One of the advances in portable communication technology is the move toward the implementation of a low intermediate frequency (IF) receiver and a direct conversion receiver (DCR). A low IF receiver converts a radio frequency (RF) signal to an intermediate frequency that is lower than the IF of a convention receiver. A direct conversion receiver downconverts a radio frequency (RF) received signal directly to baseband (DC) without first converting the RF signal to an intermediate frequency (IF). One of the benefits of a direct conversion receiver is the elimination of costly filter components used in systems that employ an intermediate frequency conversion.
A low IF or a direct conversion receiver allows the filter components to be implemented using electronic circuitry that can be located on the same device (i.e., “on-chip”) as many of the receiver components. In a direct conversion receiver implementation, high-order (e.g., fifth-order or higher) active filters are used to convert the received signal from RF to DC. Unfortunately, because the filters are implemented using electronic circuitry on the same chip as the receiver components, the filter adds significant noise to the received signal. The added noise reduces the sensitivity of the receiver, thereby making such an active filter challenging to implement.
When converting a received RF signal either to an intermediate frequency signal, or directly to a baseband signal, one or more mixers are used to downconvert the received RF signal. A mixer combines the received RF signal with a reference signal, referred to as a “local oscillator,” or “LO” signal. The resultant signal is the received signal at a different, and typically lower, frequency.
In a quadrature mixer, an in-phase signal component and a quadrature-phase signal component are presented to the mixer. At the RF frequency of the received signal, the in-phase signal component and the quadrature-phase signal component are offset in phase by a nominal 90 degrees. When implementing a mixer to downconvert a complex quadrature signal, either the RF path or the LO path must be phase-shifted to obtain the desired downconversion characteristics. For many reasons, it is desirable to phase-shift the LO signals. For example, phase-shifting the LO signals eliminates the need for costly and inefficient so-called “polyphase” filters in the RF path. Typically the LO signal is either at the RF frequency of the received signal (regular mixer) or at a fraction of the RF signal (subharmonic mixer). In a subharmonic mixer, the 90 degree phase offset between the in-phase signal component and a quadrature-phase signal component equates to a 45 degree phase offset between the LO signals when the frequency of the LO signal is one half the frequency of the RF signal.
Regardless of the mixer topology, it is desirable to efficiently generate the LO signals for a quadrature mixer.